sign in sign up
<<
Home , Desmococcus (alga), Prasiola, Prasiolaceae, Prasiola stipitata, Rosenvingiella, Trebouxiophyceae

J. Phycol. 54, 264–274 (2018) © 2018 The Authors Journal of Phycology published by Wiley Periodicals, Inc. on behalf of Phycological Society of America This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1111/jpy.12619

POLYOLS AND UV-SUNSCREENS IN THE -CLADE (, ) AS METABOLITES FOR STRESS RESPONSE AND CHEMOTAXONOMY1

Vivien Hotter, Karin Glaser Institute of Biological Sciences, Applied Ecology and Phycology, University of Rostock, Albert-Einstein-Straße 3, D-18059 Rostock, Germany Anja Hartmann, Markus Ganzera Institute of Pharmacy, Pharmacognosy, University of Innsbruck, Innrain 80-82/IV, A-6020 Innsbruck, Austria and Ulf Karsten 2 Institute of Biological Sciences, Applied Ecology and Phycology, University of Rostock, Albert-Einstein-Straße 3, D-18059 Rostock, Germany

In many regions of the world, aeroterrestrial of the Trebouxiophyceae (Chlorophyta) In contrast to their aquatic relatives, aeroterres- represent very abundant soil microorganisms, and trial algae are directly exposed to the atmosphere hence their is crucial to investigate their and thereby subject to harsh environmental condi- physiological performance and ecological impor- tions, such as strong diurnal and seasonal changes tance. Due to a lack in morphological features, of ultraviolet radiation (UVR; Hartmann et al. taxonomic and phylogenetic studies of Treboux- 2016) and strong differences between cellular and iophycean algae can be a challenging task. Since atmospheric water potential (Holzinger and Karsten chemotaxonomic markers could be a great assistance 2013). Differences in water potential drive water in this regard, 22 strains of aeroterrestrial Treboux- movement across membranes, between different cel- iophyceae were chemically screened for their polyol- lular compartments, and between organisms and patterns as well as for mycosporine-like amino acids their environment (e.g., Larcher 2003), and thus (MAAs) in their aqueous extracts using RP-HPLC and determine, for example, water availability for poik- LC-MS. D-sorbitol was exclusively detected in ilohydric organisms, such as green algae (Kranner members of the family. The novel MAA et al. 2008). Due to the incapability of green algae prasiolin and a related compound (“prasiolin-like”) to actively regulate the water budget, their cells des- were present in all investigated members of the iccate if the extracellular water potential is lower Prasiola-clade, but missing in all other tested than the intracellular one. Since life without water Trebouxiophyceae. While prasiolin could only be is impossible, uncontrolled dehydration leads to detected in field material directly after extraction, the increasing mortality unless an organism is desicca- “prasiolin-like” compound present in the other algae tion tolerant. In arid regions water is scarce and its was fully converted into prasiolin after 24 h. These availability unpredictable. Nevertheless, many mem- findings suggest D-sorbitol and prasiolin-like com- bers of the Chlorophyta and such as pounds are suitable chemotaxonomic markers for the Interfilum, Klebsormidium, , ,or Prasiolaceae and Prasiola-clade, respectively. Addi- Prasiola are found in various terrestrial habitats all tional UV-exposure experiments with selected strains over the world, from deserts and alpine regions to show that MAA formation and accumulation can be urban areas, where they inhabit both natural and induced, supporting their role as UV-sunscreen. artificial surfaces such as soil, tree bark, roof tiles, etc. (e.g., Karsten et al. 2007a,b,c, Rindi 2007, Key index words: chemotaxonomy; MAAs; polyols; Moniz et al. 2012, Darienko et al. 2015, Rysanek prasiolin; sunscreen; terrestrial algae; UV radiation et al. 2015, and references therein). Abbreviations: MAAs, mycosporine-like amino acids; Aeroterrestrial green algae developed numerous RP-HPLC, reversed phase high performance liquid mechanisms to survive desiccation (for review see chromatography; UVR, ultraviolet radiation Holzinger and Karsten 2013). Green algal members of the class Trebouxiophyceae are capable of synthe- 1Received 24 July 2017. Accepted 7 January 2018. First Published sizing and accumulating polyols. These low molecu- Online 18 January 2018. Published Online 21 February 2018, Wiley lar weight carbohydrates exhibit multiple functions. Online Library (wileyonlinelibrary.com). They act as antioxidants, stabilize proteins under 2Author for correspondence: e-mail [email protected]. Editorial Responsibility: W. Henley (Associate Editor) heat stress conditions and are rapidly available

264 CHEMOTAXONOMY IN THE PRASIOLA-CLADE 265 respiratory substrates in case of energy deficiency , but also proved its inducibility by (Yancey 2005, Karsten et al. 2007c). Polyols are also UV exposure. A chemical screening of various mem- osmotically active, that is, they decrease the intracel- bers of the Trebouxiophyceae confirmed the occur- lular water potential when accumulated (Holzinger rence of this 324 nm-MAA in spp. and and Karsten 2013). Thereby, desiccation is reduced Prasiola spp. (Karsten et al. 2005, 2007b). Recently, or even prevented without negatively affecting meta- Hartmann et al. (2016) elucidated the chemical bolic integrity. Hence, polyols are also called com- structure of this putative 324 nm-MAA in the closely patible solutes (Yancey 2005). Aeroterrestrial related Prasiola calophylla as N-[5,6 hydroxy-5(hydro- members of the Trebouxiophyceae, such as Apatococ- xymethyl)-2-methoxy-3-oxo-1-cycohexen-1-yl] glu- cus, Chloroidium, Coccomyxa, Prasiola, Rosenvingiella, tamic acid, which indeed represents a novel MAA. It and Trentepohlia, synthesize a variety of was named prasiolin. However, so far only a few polyols, such as arabitol, erythritol, glycerol, ribitol, members of the Trebouxiophyceae have been stud- D-sorbitol and volemitol (Feige and Kremer 1980, ied for the presence of this and other MAAs and Gustavs et al. 2010, 2011). While in some clades like until now the occurrence of prasiolin is experimen- Apatococcus spp. a combination of these substances tally proven only in P. calophylla. can be detected, others like Prasiola spp. contain Stress metabolites such as polyols and MAAs are only one compound (Gustavs et al. 2011). There- not only essential for the long-term survival of green fore, polyol and other low molecular weight carbo- algae under atmospheric conditions, but they can hydrate patterns can be used for chemotaxonomy also be useful in chemotaxonomy. Many aeroterres- (Karsten et al. 1999). trial green algae resemble each other morphologi- Throughout the day or season, exposure to UVR cally (Rindi 2007) and some species even display can change rapidly. While UV-C (200–280 nm) is high phenotypic plasticity (Rindi and Guiry 2002, biologically irrelevant as this wavelength is absorbed Darienko et al. 2015, 2016). This makes green algal by the ozone layer of the stratosphere, both UV-A identification down to the species level based on (315–400 nm) and UV-B (280–315 nm) reach earth’ morphological traits often complicated and some- surface. UV-B radiation is especially harmful to times even unreliable (John and Maggs 1997, Rindi many biological processes (McKenzie et al. 2007, 2007). Polyphasic approaches combining morpho- and references therein), such as photosynthesis or logical, molecular and/or physiological/biochemical enzyme activity (Holzinger and Lutz€ 2006, Sharma data sets are a promising solution to overcome these et al. 2017, and references therein). Since photosyn- problems in species identification (Proeschold and thesis is a physiological key process for algae, its Leliaert 2007, Coesel and Krienitz 2008, Darienko function is vital. et al. 2010). To oppose UVR damage, aeroterrestrial Treboux- In chemotaxonomy, chemical traits are used to iophyceae belonging to the -, Watanabea- assign organisms to taxa with equal compounds. and Prasiola-clade biosynthesize and accumulate Any chemical compound is suitable as a chemotaxo- mycosporine-like amino acids (MAAs; Karsten et al. nomic marker if it is taxon specific, consistent 2005, 2007b). These sunscreen compounds absorb within a lineage and present in detectable amounts UVR and re-emit it as harmless heat, thereby shield- (Karsten et al. 2007a). Prominent examples are pho- ing intracellular structures and biomolecules (Ban- tosynthetic pigments for algae subdivision (Roy daranayake 1998). MAAs are the most common et al. 2011) or low-molecular weight carbohydrate photoprotective compounds in aquatic organisms, patterns to distinguish lineages within the Bangio- from cyanobacteria and algae to invertebrates and phyceae (Karsten et al. 1999). Stress metabolites can fish (Dunlap and Shick 1998, Sommaruga and Gar- also be suitable chemotaxonomic markers (Dar- cia-Pichel 1999). While MAAs have been investigated ienko et al. 2010). Gustavs et al. (2011) screened a extensively in (Karsten et al. 1998, Frank- wide range of Trebouxiophyceae for their polyols lin et al. 1999, Karsten and Wiencke 1999, Karsten and detected several clade-specific patterns. For 2000, Kr€abs et al. 2002, Boedeker and Karsten 2005, instance, D-sorbitol was proposed as a marker for Pandey et al. 2017) as well as in cyanobacteria and the Prasiola-, D-ribitol for the Watanabea- and a com- (Garcia-Pichel et al. 1993, Budel et al. 1997, bination of D-ribitol and erythritol for the Apatococ- Pattanaik et al. 2008, Hartmann et al. 2015), only cus-clade (Gustavs et al. 2011). little is known about their presence in aeroterres- In this study, 22 strains of aeroterrestrial Treboux- trial Trebouxiophyceae and even less in the Prasiola- iophyceae were examined for their polyol patterns clade (Hartmann et al. 2015). A putative MAA and the presence of MAAs using HPLC. All strains within the Trebouxiophyceae was first found in Pra- were chosen due to their abundance in terrestrial siola crispa ssp. antarctica by Hoyer et al. (2001). habitats, such as biofilms or soil (Jacob et al. 1991, Using High Performance Liquid Chromatography Rindi and Guiry 2004) with a focus on species (HPLC), a unique UV-absorbing compound with an related to P. calophylla (Hartmann et al. 2016). Lat- absorption maximum at 324 nm was described. est green algal phylogenies were used as a reference Groeniger and Haeder (2002) not only confirmed (Hallmann et al. 2016, Hodac et al. 2016, Garrido- this putative 324 nm-MAA in the closely related Benavent et al. 2017, Richter et al. 2017). Based on 266 VIVIEN HOTTER ET AL. the findings of Gustavs et al. (2011), we hypothe- TABLE 1. Prasiolin/“prasiolin-like,” D-sorbitol and D-ribitol sized D-sorbitol to be a suitable chemotaxonomical contentÀ detected in the investigated Trebouxiophyceae in mg Á g 1 dry weight (DW). All investigated members of marker for the Prasiola-clade. Since UVR is a regular the Prasiola-clade are marked in bold. n.t., no trace of stressor for aeroterrestrial green algae and because respective compound. prasiolin was recently identified in P. calophylla (Hartmann et al. 2016), this and chemically similar Prasiolin/ MAAs were expected in all members of the Prasiola- prasiolin-like D-ribitol D-sorbitol Á À1 Á À1 Á À1 clade. Additionally, UVR exposure experiments were (mg g DW) (mg g DW) (mg g DW) conducted for some selected Trebouxiophyceae Prasiococcus 2.44 n.t. 75.4 strains to test the induction of MAAs as a UV pro- calcarius SAG 10.95 tective mechanism. Prasiola stipitata 12.30 15.4 31.2 SRN 124 Prasiola stipitata 21.51 11.3 27.6 MATERIALS AND METHODS SRN 125 ramosa 57.8 n.t. 87.1 Algal material and culture conditions. A total of 22 aeroter- SAG 26.83 restrial Trebouxiophyceae strains were chemically screened Trichophilus 2.53 n.t. 53.8 for the presence of both polyols and MAAs: 16 unialgal cul- welckeri SAG tures from the Sammlung von Algenkulturen at the Univer- 84.81 sity of Gottingen,€ Germany (SAG), two unialgal cultures from Rosenvingiella 10.26 n.t. 92.0 the Station Biologique de Roscoff, France, and four field sam- radicans SBDN ples (Table S1 in the Supporting Information). Strains from 005 Roscoff were grown in Provasoli-enriched full-strength seawa- Rosenvingiella 4.54 n.t. 101.5 ter (Provasoli 1968) at 13°C for 10 weeks, irradiated with radicans SBDN À À 40 lmol photons Á m 2 Á s 1 (Lumilux Cool Daylight L18W/ 1096A 865; OSRAM, Munich, Germany) under a 12:12 h light:dark Rosenvingiella 10.51 n.t. 17.2 cycle. Algal biomass was dried in silica gel prior to polyol and radicans SRN 75 MAA extraction. All SAG strains were grown in 50 mL Erlen- SAG 7.28 n.t. n.t. meyer flasks filled with modified Bold’s basal medium (Starr 43.96 and Zeikus 1993) for 16 d at a temperature of 20°C. Daylight Prasiola calophylla 2.67 n.t. 6.1 Stichococcus 0.09 n.t. 21.0 lamps (Lumilux Deluxe Daylight L15W/950; OSRAM) emit- bacillaris SAG ted PAR with a photon flux density of 25–30 lmol pho- Á À2 Á À1 397-1b tons m s under a 16:8 h light:dark cycle. Afterwards, Pseudomarvania 0.24 n.t. 44.9 algal biomass was harvested by filtration (GF 6 filters; Carl aerophytica SAG ° Roth, Karlsruhe, Germany) and dried at 30 C overnight. 2148 Prasiola calophylla field material was collected by Dr. Andreas Pseudomarvania 0.19 n.t. 21.4 Holzinger at the Botanical Garden, University of Innsbruck, ampullaeformis Austria, in January 2017 and lyophilized. Rosenvingiella radi- SAG 2047 cans (SRN 75) was collected by Dr. Svenja Heesch in Bodø, 9.48 n.t. 31.3 Norway, in October 2016 and dried in silica gel. Prasiola stipi- spinocystis SAG tata was collected by Dr. S. Heesch in Roscoff, France, in 2067 February 2017 and dried in silica gel, too (Table 1). Dry Stichococcus 0.55 n.t. 9.3 weight (DW) was always determined for all algae samples deasonii SAG prior polyol and MAA extraction. 2139 MAA induction experiment. Due to their central position Stichococcus 0.08 n.t. 10.3 jenerensis SAG within the Prasiola-clade (Hodac et al. 2016, Garrido-Benavent 2138 et al. 2017), the three strains SAG 2148, SAG 2139 and SAG 4.53 n.t. n.t. 379-1d were chosen for the UV-induction experiment. These signiensis var. isolates were pre-cultivated in 100 mL Erlenmeyer flasks for communis SAG 3 d under the conditions mentioned above to guarantee vital 2110 log phase cultures. Subsequently, the strains were transferred arboricola n.t. 13.8 n.t. to 600 mL glass petri dishes, provided with new medium and SAG 219-1a kept at 22°C–23°C for 4 d. Additionally, two radiation condi- n.t. n.t. n.t. tions were applied during a 16:8 h light:dark cycle: PAR only SAG 2007 (400–700 nm) and PAR + UVR (PAR + UV-A + UV-B, 295– bisecta n.t. n.t. n.t. 700 nm). In both control and UV-treatment, Lumilux Deluxe SAG 2043 Daylight L15W/950 (OSRAM) provided 80–90 lmol pho- Trochisciopsis n.t. n.t. n.t. À À tons Á m 2 Á s 1 PAR. UVR was emitted by Q-Panel-UVA 340 tetraspora SAG fluorescent lamps (Q-Panel, Cleveland, OH, USA). While the 19.95 control was covered with a 400 nm cut-off filter foil (Folex “Stichococcus” n.t. n.t. n.t. PR; Folex, Dreieich, Germany) resulting in total UV-A and mirabilis SAG 379- UV-B elimination, the UV-treated algal cultures were exposed 3a À À to 6–7WÁ m 2 UV-A and 0.37–0.45 W Á m 2 UV-B under a 295 nm cut-off filter (Ultraphan UBT 295; Digefra, Fursten-€ feldbruck, Germany). PAR was measured with a Li-Cor LI- 190-SB cosine corrected sensor connected to a Li-Cor LI-1000 PMA broadband radiometer (Solar Light Co., Philadelphia, data logger (Lambda Instruments, Lincoln, NE, USA). A PA, USA) was used to measure UVR. After the exposure per- iod, biomass was harvested as described above. As an CHEMOTAXONOMY IN THE PRASIOLA-CLADE 267 indicator of physiological performance, chlorophyll a fluores- version 7 (Kumar et al. 2016). Based on the lowest AIC cence, that is, maximum quantum yield of photosystem II in (Akaike 1981) calculated with jModelTest implemented in the dark-adapted state (Fv/Fm), was determined using a pulse MEGA version 7 (Kumar et al. 2016), the best evolutionary amplitude modulated fluorometer (PAM 2500; Walz, Effel- model for the data set was chosen. The phylogenetic tree was trich, Germany) according to Graiff et al. (2015). The filters constructed with the program MrBayes 3.2.2 (Ronquist et al. ° +Γ+ were dark incubated for 20 min at 22 C before Fv/Fm was 2012), using the GTR I model with 5,000,000 generations. measured (n = 5). Finally, the filters were dried and weighed Two out of four runs of Markov chain Monte Carlo were as explained above. made simultaneously, with trees taken every 500 generations. MAA analysis. The dried algal samples were ground in a Split frequencies between runs at the end of calculations microdismembrator (Sartorious, Gottingen€ Germany) in pre- were below 0.01. The trees selected before the likelihood rate cooled Teflon jars for 4 min at a shaking frequency of reached saturation were subsequently discarded. Finally, 1,800 rpm and subsequently extracted with water (100%) in branches were collapsed for clarity if the taxa were needed to an ultrasonic bath (Bandelin Sonorex 35 kHz, Berlin, Ger- obtain a stable tree but were not further investigated. many) for 30 min at 25°C. After centrifugation at 1,200g for Statistical analysis. Homoscedastic, independent t-tests 5 min (Heraeus Labofuge 400; Thermo Fisher, Waltham, were conducted with R version 3.4.0 (R-Development-Core- MA, USA), the supernatant was collected and evaporated in Team 2017). Based on the biological background and the an air stream. To guarantee exhaustive extraction, this step hypotheses, a two-tailed t-test was performed for the compar- was repeated twice. For HPLC analysis, the combined extract ison of the maximum quantum yields, whereas a one-tailed t- was re-dissolved in 5 mL water. MAA analysis was carried out test was used for the comparison of the prasiolin contents. on an Agilent 1260 HPLC system (Santa Clara, CA, USA) cou- pled to an amaZon iontrap mass spectrometer (Bruker, Bre- 9 men, Germany) using a Triart C18 column (150 3.00 mm, RESULTS 3 lm) from YMC (Dinslaken, Germany). The mobile phase was comprised of 0.25% (v/v) formic acid in water (A) and Phylogenetic analysis. To visually support the inter- 0.25% (v/v) formic acid in methanol (B). Elution was carried pretation of the results, a phylogenetic tree includ- out in isocratic mode at 2% B for 15 min, and a gradient elu- ing the 22 investigated Trebouxiophyceae was tion to 30% from 15 to 25 min, followed by a 10 min step of derived from 18S rRNA gene sequences. The re-equilibration. The DAD was set to 320 nm and flow rate, injection volume and column thermostat were adjusted to sequences of the /- and Watan- À 0.3 mL Á min 1,5lL and 30°C, respectively. Sample quantifi- abea-clade were needed to obtain a stable tree. How- cation was carried out using HPLC-UV, but for most samples ever, no member of these taxa was investigated in LC-MS experiments were additionally performed. MS-spectra this study, which is why the branches were collapsed were recorded in positive ESI mode, with a drying gas tem- for more clarity (Fig. 1). The tree mainly coincides perature of 200°C, the nebulizer gas (nitrogen) set to 23 psi, À with the phylogeny presented in Hallmann et al. and a nebulizer flow (nitrogen) of 8 L Á min 1. (2016) and Garrido-Benavent et al. (2017). More- For the quantitative determination of MAAs, a calibration curve for prasiolin was established (regression equation: over, “Stichococcus” mirabilis SAG 379-3a is located Y = 5.085 x À 2.3887; determination coefficient = 0.9999; lin- outside the Prasiola-clade, as already shown in À ear range = 1.15 to 147.6 lg Á mL 1). The second MAA Mikhailyuk et al. (2008). (“prasiolin-like”) was quantified accordingly. Polyol and MAA analysis. The polyols D-sorbitol Polyol analysis. At least 10 mg DW of both cultured and and D-ribitol were present in 16 of the 22 studied field sampled algae were used. The material was placed in aeroterrestrial Trebouxiophyceae (Table 1). An screw-capped centrifugation tubes filled with 1 mL 70% etha- exemplary chromatogram for both compounds is nol (HPLC-grade, v/v) and kept in a water bath at 70°C for 4 h. For higher extraction success, the tubes were vortexed shown in Figure 2. Except for Prasiola crispa and occasionally. After centrifugation at 13,000g for 5 min, Pseudochlorella signiensis var. communis, D-sorbitol was 800 lL of the supernatant were transferred to a new vial and found in all tested members of the Prasiola-clade evaporated to dryness under vacuum (Savant SpeedVac SVC (Hallmann et al. 2016): Prasiococcus calcarius, P. stipi- 100H; Thermo Fisher Scientific). The pellets were re-dis- tata, P. calophylla, Rosenvingiella radicans, Prasiolopsis l solved in 800 L ddH2O (HPLC-grade) and centrifuged at ramosa, Trichophilus welckeri, Pseudomarvania aerophyt- 13,000g for 5 min. The supernatant was transferred to a HPLC-vial. For analysis, an Agilent 1260 Infinity Series HPLC ica, Pseudomarvania ampullaeformis, Stichococcus bacil- system (Agilent) equipped with a vacuum degasser, a quater- laris, Stichococcus deasonii, Stichococcus jenerensis and Desmococcus spinocystis. Concentrations ranged from nary pump, a refractive index detector and a Fast Carbohy- À À drate Analysis Column (Bio Rad, Hercules, CA, USA) was ~9mgÁ g 1 DW in S. deasonii to over 100 mg Á g 1 used. Samples were eluted in ddH2O (HPLC-grade) at a flow DW in R. radicans (SBDN 1096A). Field material of Á À1 ° rate of 1 mL min at 70 C according to Karsten et al. R. radicans (SRN75) contained less D-sorbitol than (1991). cultivated material (SBDN 1096A, SBDN 005) by ~6- Phylogenetic analysis. All phylogenetic analyses are based on Á À1 18S rRNA gene sequences originating from GenBank, where fold. Around 30 mg g DW of D-ribitol were all sequences of SAG strains are publicly available. For the detected in both field samples of P. stipitata (Prasi- new isolates of Prasiola stipitata, P. calophylla and Rosenvingiella ola-clade) as an additional polyol to D-sorbitol radicans, the sequence of the most similar respective relative (Table 1). In the phylogenetically distinct Trebouxia was chosen (for full list of organisms and accession numbers arboricola, D-ribitol was the only present polyol at a À1 see Table S1). To obtain a tree that coincides with latest Tre- concentration of ~14 mg Á g DW. Neither D-sorbi- bouxiophyceae-phylogenies, additional sequences were cho- tol nor D-ribitol was detected in P. crispa, P. signiensis sen from Hallmann et al. (2016). Multiple alignments were conducted with the Muscle algorithm implemented in MEGA var. communis, Lobosphaera incisa, Myrmecia bisecta, Trochisciopsis tetraspora and “S.” mirabilis (Table 1). 268 VIVIEN HOTTER ET AL.

FIG. 1. Maximum likelihood (ML) phylogeny based on 18S rRNA gene sequences of investigated Trebouxiophyceae as well as strains taken from Hallmann et al. (2016). Thick branches indicate an ML bootstrap support ≥0.9. Bold names indicate investigation in this study. The scale bar corresponds to 0.01 substitutions per site.

The HPLC chromatogram of P. crispa did not show observation was made when using pure methanol any polyol peak, indicating that the used material for extraction, because instead of prasiolin was in a degenerated physiological state and hence (Mr = 333) another MAA with an identical UV-spec- no polyols could be extracted. trum but a molecular mass of only 332 was found. After preliminary experiments concerning the In aqueous extracts (100% water and 25% metha- optimum extraction protocol, all strains were addi- nol), both peaks appear, “prasiolin-like” at 3.1 min tionally analyzed for the presence of the MAA prasi- and prasiolin at 8.0 min (Fig. 3). Both compounds olin. These experiments were carried out using must be of highly similar structure, because they Prasiola calophylla, the strain from which prasiolin also convert into each other; after 24 h the peak was originally isolated. They showed that after grind- area of “prasiolin-like” declines and that of prasiolin ing the cells in a dismembrator, a threefold extrac- increases to the same extent. After 48 h none of the tion for 30 min each using water as solvent is two MAAs is detectable anymore. It can be hypothe- exhaustive. To account for the instability of prasi- sized that “prasiolin-like” may contain a glutamine olin, the temperature in the sonicator was kept at residue instead of glutamic acid at the nitrogen in 25°C by constantly adding ice, and evaporation of position 3, because this would explain the mass dif- the solvent was carried out under cold airstream ference of 1 Da. The isolation of “prasiolin-like,” instead of using a rotary evaporator. An interesting even though it is challenging due to instability CHEMOTAXONOMY IN THE PRASIOLA-CLADE 269

extraction, the maximum quantum yield of photosys- tem II in the dark-adapted state (Fv/Fm) was deter- mined (Fig. 4A). In Prasiola aerophytica an Fv/Fm value of 0.50 was measured under control conditions, which decreased to 0.18 after UV treatment. In S. dea- sonii an Fv/Fm value of 0.65 was detected in the con- trol, which slightly dropped to 0.55 under UV exposure. In the S. bacillaris control, the maximum quantum yield was 0.70, and UV exposure led to a minor reduction to 0.64 (Fig. 4A). Compared to the respective control, both P. aerophytica and S. deasonii showed a significantly lower Fv/Fm in the UV treat- ment (P < 0.01). Maximum quantum yield in S. bacil- = laris was also significantly reduced (t-test, t4 6.213, P < 0.05) under UVR compared to PAR (Fig. 4A). MAA inducibility and accumulation due to UV exposure was observed in all three strains (Fig. 4B). In both Stichococcus deasonii and S. bacillaris the amount of prasiolin like compounds significantly À increased from less than 0.5 mg Á g 1 DW in the con- À trol to ~5mgÁ g 1 DW in the UV treatment (t-test, = À < = À t4 4.689, P 0.01 [S. deasonii] and t4 2.583, P < 0.05 [S. bacillaris]). Although Prasiola aerophytica doubled its MAA content from less than À À 0.16 mg Á g 1 DW to ~0.3 mg Á g 1 DW, this increase = À > was not significant (t-test, t4 1.836, P 0.05; Fig. 4B).

DISCUSSION The morphological similarity and high pheno- typic plasticity of green microalgae (Rindi and Guiry 2002, Rindi 2007, Darienko et al. 2015, 2016) makes FIG. 2. HPLC chromatograms of ethanolic extracts of (A) Rosenvingiella radicans and (B) Trebouxia arboricola. While the their taxonomy a challenging task (John and Maggs R. radicans extract shows a high D-sorbitol peak at 6.7 min, a D- 1997, Rindi 2007). Moreover, in the modern molec- ribitol peak at 4.2 min was detected in T. arboricola. Abbrevia- ular age of science, morphology gets much less tions: RI signal, Refractive Index signal; rRIU, relative Refractive attention compared to former times. Instead, many Index Units. authors follow polyphasic approaches in which mor- phological data are combined with those derived reasons, is currently in progress. For this study, both from ecophysiology and cell biology as well as vari- MAAs were quantified together as sum using the cal- ous molecular markers (e.g., Darienko et al. 2010). ibration data of prasiolin (Fig. 3), and their pres- In addition, the discovery of suitable chemotaxo- ence/absence is shown in Table 1. Most interesting nomic markers might be very useful in green is the observation that prasiolin could only been microalgal taxonomy, but so far only few studies detected in Prasiola calophylla and both P. stipitata have been published on this topic (e.g., Gustavs samples, all of which were collected in the field. In et al. 2011, and references therein). Chemotaxo- all other algal strains tested, which originated from nomic characters such as polyol and MAA patterns a culture collection, “prasiolin-like” was the only are particularly helpful when sequence information MAA present (Table 1). is not available or questionable. Additionally, both Prasiolin and/or “prasiolin-like” were found in all polyol and MAA analyses are quite easy and quick tested members of the Prasiola-clade, with concen- to undertake in the lab and provide a chemical fin- À trations ranging from as low as 10 lg Á g 1 DW gerprint as part of a polyphasic approach. À (S. bacillaris)to~58 mg Á g 1 DW (P. ramosa SAG For this study, 22 aeroterrestrial Trebouxio- 26.83). Outside the Prasiola-clade, no MAAs were phyceae were chemically screened for the presence detected (Table 1). of the polyols D-sorbitol and D-ribitol as well as MAA induction experiment. To evaluate whether the MAAs, with a focus on relatives of Prasiola calophylla. UV-absorbing MAAs are inducible and accumulate Additionally, a phylogenetic tree based on 18S rRNA under controlled UVR, UV exposure experiments gene sequences was calculated (Fig. 1). It strongly were conducted with Prasiola aerophytica, Stichococcus resembles the phylogeny presented by Hallmann deasonii and S. bacillaris (Fig. 4). Prior to MAA et al. (2016) and Garrido-Benavent et al. (2017). 270 VIVIEN HOTTER ET AL.

FIG. 3. Analytical results for Prasiola calophylla. (A) LC-MS chromatogram of a freshly prepared aqueous extract and (B) the same extract analyzed after 24 h. The segments below show the MS- (3) and UV spectra (5) of the Prasiolin-like constituent, (4) and (6) the corresponding data for Prasiolin. MS spectra were recorded in positive ESI mode.

FIG. 4. Results of the MAA induction experiment under UV radiation conditions for three SAG strains from the Prasiola- clade. (A) Maximum quantum yield of PSII and (B) prasiolin/ “prasiolin-like” content in À mg Á g 1 DW after four d of PAR and UVR exposure, respectively (n = 3). Error bars indicate standard errors. *P < 0.05, **P < 0.01.

Moreover, the Prasiola-clade matches the distribution tested strains belong to the family Prasiolaceae, of D-sorbitol and prasiolin/“prasiolin-like” within the whereas P. signiensis var. communis does not (Dar- Trebouxiophyceae: The polyol D-sorbitol was exclu- ienko et al. 2016). This circumstance indicates that sively found in members of the Prasiola-clade the distribution of D-sorbitol within the Prasiola-clade (Table 1; Roser et al. 1992, Gustavs et al. 2011). The is restricted to the family Prasiolaceae. Gustavs et al. compound is absent only in one of its tested mem- (2011) chemically investigated the presence of vari- bers: Prasiola signiensis var. communis. According to ous polyols in 34 mainly aeroterrestrial Trebouxio- Garrido-Benavent et al. (2017), the 16 positively phyceae belonging to 5 different clades. The results CHEMOTAXONOMY IN THE PRASIOLA-CLADE 271 of these authors are in accordance with this study. present in any member of the Prasiola-clade. More- Considering the obvious phylogenetic distribution of over, Gustavs et al. (2011) examined P. stipitata field D-sorbitol within aeroterrestrial Trebouxiophyceae material from Germany, but only found D-sorbitol. (Table 1), these findings provide strong evidence Field material is known to be prone to contamina- for D-sorbitol as a suitable chemotaxonomic marker tion, such as epiphytic algae. Since the two P. stipi- for the family Prasiolaceae. tata field samples for this study were microscopically Based on Karsten et al. (2005) and Hartmann examined to exclude contamination prior to HPLC et al. (2016), the 324 nm MAA prasiolin was analysis, and because the detected D-ribitol concen- expected in all relatives of Prasiola calophylla. trations were of the same order of magnitude as the Indeed, in all members of the Prasiola-clade either second polyol, D-sorbitol, it is highly reasonable to prasiolin and/or a “prasiolin-like” MAA was detected assume that D-ribitol is indeed synthesized by these (Table 1). As mentioned before, prasiolin was only P. stipitata strains. The presence of a set of polyols found in field samples of P. calophylla and P. crispa, has been interpreted as a biochemical trait to better while the “prasiolin-like” compound dominated all cope with fluctuating environmental stress factors other algal strains, which were provided by the SAG that come along with a terrestrial lifestyle (Gustavs culture collection. From these data, it might be pos- et al. 2011). However, under osmotic and matric sible to assume that natural environmental condi- stress, P. crispa ssp. antarctica and a phylogenetically tions with usually high insolation stimulate the related Stichococcus species (Prasiolaceae) only accu- prasiolin formation and accumulation, while long- mulated D-sorbitol (Jacob et al. 1991, Gustavs et al. term cultivation under artificial light rather leads to 2010). The occurrence of D-ribitol as an additional the production of the “prasiolin-like” compound. polyol in both P. stipitata strains suggests a unique Furthermore, this study coincides with the find- biochemical capability of this species within the ings of Bandaranayake (1998) and Karsten et al. genus Prasiola. To confirm this hypothesis, however, (2005), although the putative 324 nm MAA men- further ecophysiological studies on P. stipitata are tioned in these earlier publications was structurally required. Nevertheless, these findings are a first hint confirmed as prasiolin just recently (Hartmann et al. that this particular Prasiola-species has additional 2016). The 324 nm MAA was also found in two mem- biochemical traits that are missing in close relatives. bers of the Watanabea-clade (Karsten et al. 2005), One hypothesis of this study was that MAAs like and most probably represents prasiolin or “prasiolin- prasiolin are UV-inducible. This seems to be the like,” too, but chemical verification using MS and case only for Stichococcus bacillaris and S. deasonii NMR techniques (Hartmann et al. 2016) is still miss- (Fig. 4B). In contrast to both Stichococcus-species, ing. The presence of this 324 nm MAA in unknown the average maximum quantum yield (Fig. 4A) in green algal specimens can be used chemotaxonomi- the Prasiola aerophytica control was at least 30% lower cally, as they can either be assigned to the Watanabea- than previously reported literature values (Juneau or to the Prasiola-clade (Karsten et al. 2005). How- and Harrison 2005, Gray et al. 2007, Kang et al. ever, the occurrence of prasiolin/“prasiolin-like” 2013, Guera et al. 2016, Zhang et al. 2017), indicat- only allows for the exclusion of taxa that lack these ing that the applied cultivation methods were not MAAs and thereby to confine the remaining relation- suitable for this species. Hence, the physiological ship-possibilities, rather than the assignment to one performance in P. aerophytica was already negatively specific clade based on the sole presence of these affected under control conditions. UV exposure led UV-sunscreens. to a slight decrease of the Fv/Fm in both Stichococcus Altogether, the phylogenetic tree (Fig. 1) empha- strains (Fig. 4A). A similar response was observed in sizes the exclusive occurrence of D-sorbitol and prasi- an unspecified Stichococcus-species isolated from a olin-like compounds in the Prasiolaceae and Prasiola- building facßade (Karsten et al. 2007b). The maxi- clade, respectively. “S.” mirabilis contains neither D- mum quantum yield in P. aerophytica, however, sorbitol, nor prasiolin/“prasiolin-like” (Table 1), severely decreased after the UV treatment (Fig 4A). which supports the position of “S.” mirabilis outside The MAA content in both control and UVR the Prasiola-clade as already shown in Mikhailyuk exposed algae mirrored these findings, as only a et al. (2008). A re-evaluation of this species should minor accumulation was observed (Fig. 4B): As be considered, as both the polyol and MAA content MAAs are UV protectants (Bandaranayake 1998), as well as the phylogenetic position derived from the strong increase in MAAs in both Stichococcus 18S rRNA gene sequence show that “S.” mirabilis strains explains the relatively low impact of UVR on does not belong to the genus Stichococcus. their maximum quantum yield. Conversely, the low The polyol D-ribitol was unexpectedly detected in MAA content in P. aerophytica might be a reason for two independent Prasiola stipitata strains. Even the low maximum quantum yield after UV expo- though this polyol is widely distributed, for instance sure. Nevertheless, considering the general presence in members of the Watanabea-, - and Tre- of prasiolin and/or “prasiolin-like” in P. aerophytica bouxia-clade (Maruo et al. 1965, Richardson and (Table 1; Fig. 4B) and its phylogenetic position Smith 1968, Gustavs et al. 2010, 2011, Sadowsky within the Prasiola-clade, physiologically unaffected et al. 2016), it has not yet been reported to be P. aerophytica is most likely capable of MAA 272 VIVIEN HOTTER ET AL. accumulation, too. Nevertheless, the prasiolin/“- integrative taxonomy and DNA barcoding with further impli- siolin-like” MAAs were shown to be inducible under cations for the species identification in environmental sam- ples. PLoS ONE 10:e0127838. UVR. Thereby, the results of this part of the pre- Darienko, T., Gustavs, L., Mudimu, O., Menendez, C. R., Schu- sented study are in accordance with Karsten et al. mann, R., Karsten, U., Friedl, T. & Proschold, T. 2010. (2007b) and additionally prove that prasiolin/“- Chloroidium, a common terrestrial coccoid green alga previ- ously assigned to (Trebouxiophyceae, Chlorophyta). siolin-like” are UV-inducible, and most probably UV- – protective substances. Moreover, these findings pro- Eur. J. Phycol. 45:79 95. Darienko, T., Gustavs, L. & Proschold,€ T. 2016. Species concept vide new evidence that the distribution of these and nomenclatural changes within the genera Elliptochloris MAAs is not only attributable to phylogenetic rela- and Pseudochlorella (Trebouxiophyceae) based on an integra- tions, but also to ecophysiological acclimation. tive approach. J. Phycol. 52:1125–45. The main goal of this study was to emphasize the Dunlap, W. C. & Shick, J. M. 1998. Ultraviolet radiation-absorb- ing mycosporine-like amino acids in coral reef organisms: value of easily detectable chemical traits in green a biochemical and environmental perspective. J. Phycol. microalgal taxonomy. Furthermore, we aimed to 34:418–30. establish the polyol D-sorbitol and prasiolin/“- Feige, G. B. & Kremer, B. P. 1980. Unusual carbohydrate pattern siolin-like” as suitable chemotaxonomic markers for in Trentepohlia species. Phytochem. 19:1844–5. the Prasiolaceae and Prasiola-clade, respectively. Franklin, L. A., Yakovleva, I., Karsten, U. & Luning, K. 1999. Syn- thesis of mycosporine-like amino acids in Chondrus crispus Both are highly abundant taxa of aeroterrestrial (Florideophyceae) and the consequences for sensitivity to green algae with a worldwide distribution, but also a ultraviolet B radiation. J. Phycol. 35:682–93. challenging taxonomy. Especially D-sorbitol was pro- Garcia-Pichel, F., Wingard, C. E. & Castenholz, R. W. 1993. Evi- ven to be of high chemotaxonomic value, as it is dence regarding the UV sunscreen role of a mycosporine- like compound in the cyanobacterium Gloeocapsa sp. Appl. exclusively found in the Prasiolaceae-family. Prasi- Environ. Microb. 59:170–6. olin/“prasiolin-like” compounds are not only Garrido-Benavent, I., Perez-Ortega, S. & de los Rıos, A. 2017. present in the Prasiola-clade, but also in the Watan- From Alaska to Antarctica: species boundaries and genetic abea-clade (Karsten et al. 2005), and thus have a diversity of Prasiola (Trebouxiophyceae), a foliose chloro- wider distribution amongst the Trebouxiophyceae phyte associated with the bipolar -forming fungus Mas- todia tessellata. Mol. Phylogenet. Evol. 107:117–31. than D-sorbitol. Nevertheless, they still are a suitable Graiff, A., Liesner, D., Karsten, U. & Bartsch, I. 2015. Tempera- chemotaxonomic marker, because their presence ture tolerance of western Baltic Sea Fucus vesiculosus – reduces the remaining possible affiliations down to growth, photosynthesis and survival. J. Exp. Mar. Biol. Ecol. – either the Prasiola-, or the Watanabea-clade. Since 471:8 16. Gray, D. W., Lewis, L. A. & Cardon, Z. G. 2007. Photosynthetic aeroterrestrial green algae are taxonomically highly recovery following desiccation of desert green algae (Chloro- diverse, with many taxa belonging to the Chloro- phyta) and their aquatic relatives. , Cell Environ. phyta and Streptophyta, this study is the first to 30:1240–55. simultaneously establish two groups of chemically Groeniger, A. & Haeder, D. P. 2002. Induction of the synthesis of independent chemotaxonomic markers, a polyol an UV-absorbing substance in the green alga Prasiola stipitata. J. Photochem. Photobiol. B-Biol. 66:54–9. and MAAs, for the Prasiola-clade. In addition, fur- Guera, A., Gasulla, F. & Barreno, E. 2016. Formation of photosys- ther investigations on this highly useful and promis- tem II reaction centers that work as energy sinks in lichen ing topic are required to better understand the symbiotic Trebouxiophyceae microalgae. Photosynth. Res. – protective functions of both metabolites in these ter- 128:15 33. Gustavs, L., Eggert, A., Michalik, D. & Karsten, U. 2010. Physio- restrial green algae. logical and biochemical responses of green microalgae from different habitats to osmotic and matric stress. Protoplasma The authors deeply thank Dr. Svenja Heesch and Prof. 243:3–14. Andreas Holzinger for providing algal samples. This study Gustavs, L., Gors, M. & Karsten, U. 2011. Polyol patterns in bio- was financially supported by the Deutsche Forschungsgemein- film-forming aeroterrestrial green algae (Trebouxiophyceae, schaft (DFG; KA899/16) to U.K. and the Austrian Science Chlorophyta). J. Phycol. 47:533–7. Fund (FWF; ZFP296710) to M.G. Hallmann, C., Hoppert, M., Mudimu, O. & Friedl, T. 2016. Biodi- versity of green algae covering artificial hard substrate sur- faces in a suburban environment: a case study using Akaike, H. 1981. Citation classic - a new look at the statistical- molecular approaches. J. Phycol. 52:732–44. model identification. CC/Eng. Tech. Appl. Sci. 51:22–22. Hartmann, A., Becker, K., Karsten, U., Remias, D. & Ganzera, M. Bandaranayake, W. M. 1998. Mycosporines: are they nature’s sun- 2015. Analysis of mycosporine-like amino acids in selected screens? Nat. Prod. Rep. 15:159–72. algae and cyanobacteria by hydrophilic interaction liquid Boedeker, C. & Karsten, U. 2005. The occurrence of mycosporine- chromatography and a novel MAA from the red alga Cate- like amino acids in the gametophytic and sporophytic stages nella repens. Mar. Drugs 13:6291–305. of Bangia (Bangiales, Rhodophyta). Phycologia 44:403–8. Hartmann, A., Holzinger, A., Ganzera, M. & Karsten, U. 2016. Budel, B., Karsten, U. & Garcia-Pichel, F. 1997. Ultraviolet-absorb- Prasiolin, a new UV-sunscreen compound in the terrestrial ing scytonemin and mycosporine-like amino acid derivatives green macroalga Prasiola calophylla (Carmichael ex Gre- in exposed, rock-inhabiting cyanobacterial lichens. Oecologia ville) Kutzing (Trebouxiophyceae, Chlorophyta). Planta 112:165–72. 243:161–9. Coesel, P. F. M. & Krienitz, L. 2008. Diversity and geographic dis- Hodac, L., Hallmann, C., Spitzer, K., Elster, J., Fasshauer, F., tribution of desmids and other coccoid green algae. Biodiver. Brinkmann, N., Lepka, D., Diwan, V. & Friedl, T. 2016. Conserv. 17:381–92. Widespread green algae Chlorella and Stichococcus exhibit Darienko, T., Gustavs, L., Eggert, A., Wolf, W. & Proschold, T. polar-temperate and tropical-temperate biogeography. FEMS 2015. Evaluating the species boundaries of green microalgae Microb. Ecol. 92:fiw122. (Coccomyxa, Trebouxiophyceae, Chlorophyta) using CHEMOTAXONOMY IN THE PRASIOLA-CLADE 273

Holzinger, A. & Karsten, U. 2013. Desiccation stress and toler- Larcher, W. 2003. Physiological Plant Ecology: Ecophysiology and Stress ance in green algae: consequences for ultrastructure, Physiology of Functional Groups. Springer-Verlag, Berlin, Hei- physiological, and molecular mechanisms. Front. Plant Sci. delberg, 514 pp. 4:327. Maruo, B., Hattori, T. & Takahashi, H. 1965. Excretion of ribitol Holzinger, A. & Lutz,€ C. 2006. Algae and UV irradiation: effects and sucrose by green algae into the culture medium. Agr. on ultrastructure and related metabolic functions. Micron Biol. Chem. 29:1084–9. 37:190–207. McKenzie, R. L., Aucamp, P. J., Bais, A. F., Bjorn,€ L. O. & Ilyas, Hoyer, K., Karsten, U., Sawall, T. & Wiencke, C. 2001. Photopro- M. 2007. Changes in biologically-active ultraviolet radiation tective substances in Antarctic macroalgae and their variation reaching the Earth’s surface. Photochem. Photobiol. Sci. 6:218– with respect to depth distribution, different tissues and 31. developmental stages. Mar. Ecol. Prog. Ser. 211:117–29. Mikhailyuk, T. I., Sluiman, H. J., Massalski, A., Mudimu, O., Dem- Jacob, A., Kirst, G. O., Wiencke, C. & Lehmann, H. 1991. chenko, E. M., Kondratyuk, S. Y. & Friedl, T. 2008. New Physiological responses of the Antarctic green alga Prasiola Streptophyte green algae from terrestrial habitats and an crispa ssp. antarctica to salinity stress. J. Plant Physiol. 139:57–62. assessment of the genus Interfilum (Klebsormidiophyceae, John, D. M. & Maggs, C. 1997. Species problems in eukaryotic Streptophyta). J. Phycol. 44:1586–603. algae: a modern perspective. In Claridge, M. F., Dawah, H. Moniz, M. B. J., Rindi, F. & Guiry, M. D. 2012. Phylogeny and tax- A. & Wilson, M. R. [Eds.] Species: The Units of Biodiversity. onomy of (Trebouxiophyceae, Chlorophyta) from Chapman & Hall, London, pp. 83–107. Tasmania, including Rosenvingiella tasmanica sp. nov. Phycolo- Juneau, P. & Harrison, P. J. 2005. Comparison by PAM fluorome- gia 51:86–97. try of photosynthetic activity of nine marine phytoplankton Pandey, A., Pandey, S., Pathak, J., Ahmed, H., Singh, V., Singh, S. grown under identical conditions. Photochem. Photobiol. P. & Sinha, R. P. 2017. Mycosporine-like amino acids (MAAs) 81:649–53. profile of two marine red macroalgae, Gelidium sp. and Cera- Kang, E. J., Scrosati, R. A. & Garbary, D. J. 2013. Physiological mium sp. Inter. J. Appl. Sci. Biotech. 5:12–21. ecology of photosynthesis in Prasiola stipitata (Trebouxio- Pattanaik, B., Roleda, M. Y., Schumann, R. & Karsten, U. 2008. phyceae) from the Bay of Fundy, Canada. Phycol. Res. Isolate-specific effects of ultraviolet radiation on photosyn- 61:208–16. thesis, growth and mycosporine-like amino acids in the Karsten, U., West, J. A., Zuccarello, G. C., Nixdorf, O., Barrow, K. microbial mat-forming cyanobacterium Microcoleus chthono- D. & King, R. J. 1999. Low molecular weight carbohydrate plastes. Planta 227:907–16. patterns in the (Rhodophyta). J. Phycol. Proeschold, T. & Leliaert, F. 2007. Systematics of the green 35:967–76. algae: conflict of classic and modern approaches. In Brodie, Karsten, U. 2000. Occurrence of photoprotective mycosporine- J. & Lewis, J. M. [Eds.] Unravelling the Algae: The Past, Pre- like amino acid compounds (MAAs) in marine red macroal- sent, and Future of Algal Systematics. The Systematics Associa- gae from temperate Australian waters. Proc. Linn. Soc. N. S. tion Special Volume Series, 75. Taylor and Francis, London, W. 122:123–9. pp. 123–53. Karsten, U., Friedl, T., Schumann, R., Hoyer, K. & Lembcke, S. Provasoli, L. 1968. Media and prospects for the cultivation of 2005. Mycosporine-like amino acids and phylogenies in marine algae. In Watanabe, A. & Hattori, A. [Eds.] Cultures green algae: Prasiola and its relatives from the Trebouxio- and Collections of Algae. Japanese Society Plant Physiology, phyceae (Chlorophyta). J. Phycol. 41:557–66. Hakone, pp. 63–75. Karsten, U., Gors,€ S., Eggert, A. & West, J. A. 2007a. Trehalose, R-Development-Core-Team. 2017. R: A Language and Environment digeneaside, and floridoside in the Florideophyceae (Rhodo- for Statistical Computing. R Foundation for Statistical Comput- phyta) - a reevaluation of its chemotaxonomic value. Phycolo- ing, Vienna, Austria. gia 46:143–50. Richardson, D. H. S. & Smith, D. C. 1968. Lichen physiology. IX. Karsten, U., Karsten, U., Lembcke, S. & Schumann, R. 2007b. Carbohydrate movement from the Trebouxia symbiont of Xan- The effects of ultraviolet radiation on photosynthetic perfor- thoria aureola to the fungus. New Phytol. 67:61–8. mance, growth and sunscreen compounds in aeroterrestrial Richter, D., Matuła, J., Urbaniak, J., Waleron, M. & Czerwik- biofilm algae isolated from building facades. Planta 225:991– Marcinkowska, J. 2017. Molecular, morphological and ultra- 1000. structural characteristics of Prasiola crispa (Lightfoot) Kutz-€ Karsten, U., Sawall, T., Hanelt, D., Bischof, K., Figueroa, F. L., ing (Chlorophyta) from Spitsbergen (Arctic). Polar Biol. Flores-Moya, A. & Wiencke, C. 1998. An inventory of UV- 40:379–97. absorbing mycosporine-like amino acids in macroalgae from Rindi, F. 2007. Diversity, distribution and ecology of green algae polar to warm-temperate regions. Bot. Mar. 41:443–53. and cyanobacteria in urban habitats. In Seckbach, J. [Ed.] Karsten, U., Schumann, R. & Mostaert, A. 2007c. Aeroterrestrial Algae and Cyanobacteria in Extreme Environments. Springer algae growing on man-made surfaces. In Seckbach, J. [Ed.] Netherlands, Dordrecht, pp. 619–38. Algae and Cyanobacteria in Extreme Environments. Springer Rindi, F. & Guiry, M. D. 2002. Diversity, life history, and ecology Netherlands, Dordrecht, pp. 583–97. of Trentepohlia and Printzina (Trentepohliales, Chlorophyta) Karsten, U., Thomas, D. N., Weykam, G., Daniel, C. & Kirst, G. in urban habitats in western Ireland. J. Phycol. 38:39–54. O. 1991. A simple and rapid method for extraction and sepa- Rindi, F. & Guiry, M. D. 2004. Composition and spatial variability ration of low molecular weight carbohydrates from macroal- of terrestrial algal assemblages occurring at the bases of gae using high-performance liquid chromatography. Plant urban walls in Europe. Phycologia 43:225–35. Physiol. Biochem. 29:373–8. Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Dar- Karsten, U. & Wiencke, C. 1999. Factors controlling the forma- ling, A., Hohna, S., Larget, B., Liu, L., Suchard, M. A. & tion of UV-absorbing mycosporine-like amino acids in the Huelsenbeck, J. P. 2012. MrBayes 3.2: efficient Bayesian phy- marine red alga Palmaria palmata from Spitsbergen (Norway). logenetic inference and model choice across a large model J. Plant Physiol. 155:407–15. space. Syst. Biol. 61:539–42. Kr€abs, G., Bischof, K., Hanelt, D., Karsten, U. & Wiencke, C. 2002. Roser, D. J., Melick, D., Ling, H. & Seppelt, R. 1992. Polyol and Wavelength-dependent induction of UV absorbing mycospor- sugar content of terrestrial from continental Antarc- ine-like amino acids in the red alga Chondrus crispus under nat- tica. Antarct. Sci. 4:413–20. ural solar radiation. J. Exp. Mar. Biol. Ecol. 268:69–82. Roy, S., Llewellyn, C. A., Egeland, E. S. & Johnsen, G. 2011. Phyto- Kranner, I., Beckett, R., Hochman, A. & Nash, T. H. 2008. Desic- plankton Pigments: Characterization, Chemotaxonomy and Applica- cation-tolerance in lichens: a review. Bryologist 111:576–93. tions in Oceanography. Cambridge University Press, Kumar, S., Stecher, G. & Tamura, K. 2016. MEGA7: Molecular Cambridge, 845 pp. Evolutionary Genetics Analysis Version 7.0 for Bigger Data- Rysanek, D., Hrckova, K. & Skaloud, P. 2015. Global ubiquity and sets. Mol. Biol. Evol. 33:1870–4. local endemism of free-living terrestrial protists: 274 VIVIEN HOTTER ET AL.

phylogeographic assessment of the streptophyte alga Kleb- Zhang, P., Li, Z., Lu, L. H., Xiao, Y., Liu, J., Guo, J. S. & Fang, F. sormidium. Environ. Microb. 17:689–98. 2017. Effects of stepwise nitrogen depletion on carotenoid Sadowsky, A., Mettler-Altmann, T. & Ott, S. 2016. Metabolic content, fluorescence parameters and the cellular stoichiom- response to desiccation stress in strains of green algal photo- etry of . Spectrochim. Acta A 181:30–8. bionts (Trebouxia) from two Antarctic lichens of southern habitats. Phycologia 55:703–14. Sharma, S., Chatterjee, S., Kataria, S., Joshi, J., Datta, S., Vairale, M. G. & Veer, V. 2017. A review on responses of plants to Supporting Information UV-B radiation related stress. In Singh, V. P., Singh, S., Pra- sad, S. M. & Parihar, P. [Eds.] UV-B Radiation: From Environ- Additional Supporting Information may be mental Stressor to Regulator of Plant Growth. John Wiley & Sons Ltd, Hoboken, New Jersey, pp. 75–97. found in the online version of this article at the Sommaruga, R. & Garcia-Pichel, F. 1999. UV-absorbing mycospor- publisher’s web site: ine-like compounds in planktonic and benthic organisms from a high-mountain lake. Arch. Hydrobiol. 144:255–69. Table S1. List of algae used for phylogenetic Starr, R. C. & Zeikus, J. A. 1993. UTEX - the culture collection of analysis. algae at the University of Texas at Austin 1993 list of cul- tures. J. Phycol. 29:1–106. Yancey, P. H. 2005. Organic osmolytes as compatible, metabolic and counteracting cytoprotectants in high osmolarity and other stresses. J. Exp. Biol. 208:2819–30.